Return-to-Work Outcomes Among Social Security Disability Insurance (DI) Beneficiaries Yonatan Ben-Shalom Arif Mamun Presented at the CSDP Forum Washington, DC September 17, 2014
Acknowledgments The research reported herein was performed pursuant to a grant from the National Institute on Disability and Rehabilitation Research (NIDRR) and funded as part of the Employment Policy and Measurement Rehabilitation Research and Training Center (H133B100030) The opinions and conclusions expressed are solely those of the authors and do not represent the opinions or policy of NIDRR or any agency of the federal government 2
Motivation Policymakers are interested in promoting employment among DI beneficiaries Targeting specific groups of DI beneficiaries is likely of interest Recent initiatives targeted to specific groups include the Mental Health Treatment Study, Accelerated Benefits, YTD, and PROMISE Policymakers need more information on what factors are associated with return-to-work 3
How are beneficiary characteristics and local economic conditions associated with return-to-work outcomes? We specifically focus on Age and impairment type State economies and other state-level effects Timing of award Research Question We present a long-term view of what factors influence outcomes among DI beneficiaries Five years after DI award Most beneficiaries who return to work do so within five years 4
Data Sources Administrative records from Social Security Administration (SSA) Rehabilitation Services Administration (RSA) Data from SSA and RSA matched at the individual level Monthly state unemployment rates from the Bureau of Labor Statistics 5
Study Population and Study Period New working-age DI awardees from January 1996 to December 2004 Analysis sample excluded beneficiaries who died or reached age 65 within five years of award 6
Outcomes: Return-to-Work Milestones in Five Years Following DI Award Service enrollment Enrollment in services from state vocational rehabilitation agency (SVRA) or employment network (EN) Trial work period (TWP) completion Suspension or termination due to work (STW) 7
Test Your Knowledge Which DI award cohort had better 5-year return to work outcomes? a) 1996, when the economy was expanding b) 2000, at a peak of economic expansion
Share of New DI Awardees (1996 2004) Achieving Milestones Declines with Age 30% 25% Service enrollment TWP completion STW 20% 15% 10% 5% 0% 18 24 25 29 30 39 40 49 50 57 58 59 Age Group 9
Share of New DI Awardees (1996 2004) Achieving Milestones Varied by Impairment 25% 20% 15% Sensory impairments Intellectual disability Other psychiatric disorders Affective disorders Other physical disorders Other musculoskeletal disorders Back disorders 10% 5% 0% Service enrollment TWP completion STW Return to Work Milestones 10
Multivariate Analysis Differences in outcomes across impairment groups might be driven by differences in age distribution of beneficiaries across impairment groups, and vice versa To account for this, we conduct multivariate analyses We control for individual characteristics, programmatic features (including DI award month), and state effects 11
Likelihood of Achieving Milestones Likelihood of achieving milestones increases with Sensory impairments Being black Years of education DI award at initial adjudication level Likelihood of achieving milestones decreases with Age Back and other musculoskeletal disorders Monthly DI benefit Receipt of SSI benefits Medicare eligibility at award Disabled adult child status Increase in state unemployment rate 12
Large Variation Exists in Return-to-Work Outcomes by State of Residence 25% 20% TWP completion STW Service enrollment 15% 10% 5% 0% DC MA AK VT MN WA IL CA SD AZ CO IA NH NV OR NY ME NJ MD ND DE WY OH KS UT WI NM HI NE PA TX CT FL IN MI WV LA ID RI MO MS VA KY PR AR OK MT GA NC SC TN AL Note: States are ordered from largest to smallest effects in STW regression 13
DI Award Month and Return-to-Work Outcomes Award month captures unobserved factors affecting behavior among beneficiaries who enter DI at different points in time Changes in policy over time Changes in the composition of new awardees in terms of unobserved characteristics Recall that we account for beneficiary characteristics, state economy, and other fixed state conditions 14
Economic-Expansion-Period Awardees Seem Progressively Less Likely to Achieve Milestones 4% 2% Six-month moving average 0% -2% TWP Completion STW -4% Jan-96 Apr-96 Jul-96 Oct-96 Jan-97 Apr-97 Jul-97 Oct-97 Jan-98 Apr-98 Jul-98 Oct-98 Jan-99 Apr-99 Jul-99 Oct-99 Jan-00 Apr-00 Jul-00 Oct-00 Jan-01 Apr-01 Jul-01 Oct-01 Jan-02 Apr-02 Jul-02 Oct-02 Jan-03 Apr-03 Jul-03 Oct-03 Jan-04 Apr-04 Jul-04 Oct-04 Jan-05 Month of DI Award 15
Award-month influenced service enrollment differently than other milestones 4% 2% Service Enrollment Six-month moving average 0% -2% TWP Completion STW -4% Jan-96 Apr-96 Jul-96 Oct-96 Jan-97 Apr-97 Jul-97 Oct-97 Jan-98 Apr-98 Jul-98 Oct-98 Jan-99 Apr-99 Jul-99 Oct-99 Jan-00 Apr-00 Jul-00 Oct-00 Jan-01 Apr-01 Jul-01 Oct-01 Jan-02 Apr-02 Jul-02 Oct-02 Jan-03 Apr-03 Jul-03 Oct-03 Jan-04 Apr-04 Jul-04 Oct-04 Jan-05 Month of DI Award 16
Conclusions Younger beneficiaries substantially more likely to achieve return-to-work milestones Beneficiaries with sensory impairments have highest likelihood of achieving all milestones; those with back and other musculoskeletal disorders show the lowest two likelihoods Differences across age and impairment persist after accounting for other characteristics Large variation in the relationship between state of residence and return-to-work outcomes Timing of award may be capturing unobserved beneficiary characteristics 17
Policy Issues Should SSA direct efforts to help beneficiaries return to work toward recent awardees under age 40? Should return-to-work initiatives be tailored according to impairment type? Should efforts to slow entry via early intervention focus on those most likely to achieve the return-to-work milestones under current law? What factors explain variation across states and time? 18
For More Information Please contact Arif Mamun amamun@mathematica-mpr.com Yonatan Ben-Shalom yben-shalom@mathematica-mpr.com 19